ShaunR

Merry Xmas:Symlink.llb

Yeah. but it worked n 2009

I think we've articulated the goal well enough. However, a VIM doesn't necessarily have a reference like Queues et. al. so I'm not sure there would be an "uncomplicated" first order solution for it even though it's a kind of natural progression from singular VIMs to libraries of VIMs. In theory, the second order solution is to use those primitives internally since they have that behaviour which should propegate out but I was unable to see if that worked with DVRs and trying queues didn't illicit the behaviour.

Seems VIMs can no longer be used in polymorphic VIs either. I don't think this is a good idea. The VIMs I've played around with (before the type case structure was introduced) relied on at least one terminal having a type which would dictate the undefined types due to the nature of the internal polymorphic VIs (like a Queue or DVR). This is why I was trying to get a VIM to create a DVR which "should" define the type across all connected VIs which would solve this problem. In a nutshell, I'm expecting this set of VIMs (rightly or wrongly) to behave like the Queue, Notifier et. al. primitives when the Enqueue and Dequeue element terminal is unwired.

VIMs have to be inlined now therefore:

Never mind. It seems that you cannot have DVRs inside VIMs any more.

If you were to use a DVR (or the user supplied DVR) and put the reference onto a single element queue inside the ViM. Wouldn't you get all the type checking and polymorphism for free across all the VIs without all the type cases?

I can't find them in the examples finder (tried a few different words). The button on VIPM shows the directory with them, though. FWIW. I hate the example finder. You neeed to know the right words for what you're searching for and it's a lot of clicks and typing to get there (ignoring the excrutiating pain of actually creating the Example Finder special files). It's much preferable to have them in the same pallet as the VIs; as you had originally. They need to take that requirement off of the tools network checklist.

The examples were missing when I installed the VIP package so it took me a bit of time to figure out why wires were broken (see bottom row of VIs). The rows above are what happens when the correct data type is applied to the "Add To Buffer" Is it possiblee to to make the "Add To Buffer" select the correct instance depending on the "Create Buffer" data type?

It's the other way round. You include the files contained in an LLB in a lvlib. You end up with two files, the lvlib (XML description) and the LLB (the Vis).

They will be vetted and approved by their respective governments. The problem as I see it isn't limiting execution to signed code, it is who has capability to create and control the keys. "Derived keys" are weasel words for backdoors. But I see it as worse than that. These proposed hardware keys only limit what you can run, not what they (M$ et.al.) can run and they are major players in my threat model. Remember the the Clipper Chip? Or the Apple "Kill Switch"?. These kinds of technologies are packaged as "security" but give the corporate behemoths an inordinate amount of control over your devices. The technology already exists for self-signed secure boot. and a similar route for execution is far more appropriate. You don't need to trust anyone else but yourself (by that I mean your company) - you can't outsource security! Nope. I don't trust them as far as I can throw them. I go to great lengths to mitigate exfiltration of information to the likes of Google, M$ (i refuse to use Apple products) etc. I block all sites that are behind Coudflare (which is a MITM) and I use my own servers for things like Git, SVN, Jenkins etc. I realise the current trend is for online services (e.g Github) storing all your data but this is folly IMO. I may have to use a M$ OS, but I can and do limit their ability to leverage their indisdious telemetry and who has Driver Signing turned on? The annointed ones? China has it's own Certificate Authorities so good luck with that if your data gets routed through their servers. Cert Auth is a poor solution that only really works for the man-in-the-street because the root certificates are routinely distributed with browsers (Windows updates it's store once a week, IIRC). When did NI last update their users' LabVIEW CA file for the HTTPS Vis? What exactly is NI's policy on this and where is it documented?

Signing is great but just be aware the weakness is key distribution. Not so much a problem with physical distributions but a concern over the internet depending on your threat model.

Chinese, Russian and North Korean companies too? Excuse me if I'm not entralled by allowing M$/Intel/Apple/etc backdoors or control of what I can and can't run in an age where computers are always connected and OS telemetry is rife. I reject this dystopian sales pitch for "security" which is nothing more than a hardware version of Certificate Authorities aimed squarely at market control.

Hmm. Just looked at my creatfile in the Zip Library. Seems I used "FNewRefNum" rather than a cast so I was obviously talking out of my backside. I guess I remembered all the stuff that didn't work rather than what did.

Is that a consequence of using an intermediary DLL and "adapt to type" on the CLFN? Using the type cast at the LabVIEW level should mitigate this for TCP and file refs. Depends on what you want to do with them. Deleting the target? If the LabVIEW delete function is based on deletefile (and I can't remember off hand if it is because it doesn't take a filename), then deleting the target depends on the FILE_FLAG_DELETE_ON_CLOSE when calling createfile. I'll have to go back over this because I figured it out last time with the Zip library but for release, only needed long filename support (and even then, only for extraction). I added the methods to create and read the symlink attribute with a view to adding the functionality later after some bug reports. IIRC writing symlinks wasn't much of a problem, reading them were a little bit harder but shorcuts were a pain which I decided not to do. The file read and write only take a handle and don't care about most of these issues. Of course I haven't concerned myself with Mac and Linux.